first off, that heatsink round-up is very nice and all, but it is impossible to compare the heatsinks from the c/w values listed on the site.
first, they are using different fans in each test
second, the c/w values are rather crudely calculated from testbeds that are not only primitive, but also inconsistent, as listed are heatsinks from over several years, during which the testbeds have changed drastically.
personally, I would keep the mcx-4000, because it is still one of the best out there, and though thermalright's latest offerings would edge it out in performance, it's not worth the 40 bucks